How Automated Loan Approval Is Closing the Gender Gap in Small Business Finance

A World Bank-backed study in Peru finds that gender bias in SME lending arises mainly from human discretion, with women receiving smaller and less attractive loans despite similar profiles. When loan approvals are automated using psychometric scoring, these disparities disappear without increasing default risk.

In the world of small business finance, access to credit often determines whether a company grows or struggles to survive. Yet for decades, women entrepreneurs have faced subtle but persistent disadvantages when applying for loans. A new study by researchers from the World Bank's Development Research Group, IDB Invest, and the Inter-American Development Bank suggests that technology could offer a powerful solution.

The research explores a simple but important question: Can automated loan approvals reduce gender bias in lending? The answer, based on real-world evidence from Peru, appears to be yes.

A Natural Experiment in Peru's Banking Sector

The study focuses on a pilot program run by a large commercial bank in Peru. The bank introduced a psychometric credit scoring tool that evaluates applicants based on traits like personality, intelligence, and integrity, rather than relying only on financial history.

Here is how it worked. Applicants who scored above a certain cutoff were automatically approved for loans, with the loan size determined by their score. Those who scored below the cutoff were assessed through the traditional process, where loan officers reviewed applications, visited businesses, and made judgment calls.

This setup created a rare opportunity to compare two systems side by side: human decision-making versus automated approval.

Where Human Judgment Shows Its Limits

The findings reveal a clear pattern. When loan officers handled decisions, gender differences emerged. Women were just as likely as men to receive loan offers, but they were less likely to accept them.

The reason becomes clearer when looking at loan terms. Women received significantly smaller loan amounts than men, even when their businesses were similar. This suggests that the bias was not about outright rejection, but about less favorable conditions that made loans less attractive.

In simple terms, women were being offered deals that were harder to accept.

What Changes When Decisions Are Automated

The picture shifts dramatically when the algorithm takes over. Above the approval threshold, where decisions were fully automated, the gender gap disappears.

Women and men received similar loan amounts, accepted loans at similar rates, and had equal access to credit. The consistency of the algorithm removed the variation that came with human judgment.

This shows that standardizing decisions can eliminate unequal treatment, even when that bias is subtle rather than obvious.

No Trade-Off Between Fairness and Risk

One major concern about automation is whether it leads to riskier lending. If machines replace human judgment, could banks end up making poorer decisions?

The study finds no evidence of this. Loan repayment rates remained stable, and women were actually less likely to default than men in some cases. This means fairer lending did not come at the cost of higher financial risk.

In fact, the findings suggest that traditional methods may underestimate the creditworthiness of women entrepreneurs.

What This Means for the Future of Finance

The results carry important lessons for banks and policymakers. Even in countries with strong legal protections for gender equality, bias can still shape outcomes when decisions rely on human discretion.

Automation, when designed carefully, can help remove these biases by applying the same rules to everyone. However, not all algorithms are equal. Systems based on past data may still reflect old inequalities. What makes this case different is the use of psychometric data, which focuses on potential rather than history.

The takeaway is clear. Technology alone is not the answer, but it can be part of a smarter, fairer system. As banks modernize their processes, combining innovation with thoughtful design could open the door to more inclusive access to finance.

In the end, the study offers a hopeful message. With the right tools, it is possible to build lending systems that are not only more efficient but also more equitable.